Motif 3

Released
2026-08-07
Last refreshed
2026-10-06
Status
Researched 46d ago
Open sourceCommercial use: permittedCodingRAGAgentsLong contextJSON / Tool use

Motif 3 is available now for coding, rag, and agents with open-source and 256K context; evaluate it while provider pricing coverage matures.

Motif Technologies releases · 1 in the last 12 monthsChangelog →
Specifications
Family
Motif
Released
2026-08-07
Context
256K
Parameters
314B (13.2B active)
Architecture
Mixture of Experts
Specialization
general
Openness
Open source
License
MITOSI-approvedPermitted
Weights
Available
Code
Unknown
Training
Multi-stage
Pricing

No tracked provider pricing is available yet.

About

Motif 3 (HF Motif-Technologies/Motif-3) is Motif Technologies' decoder-only mixture-of-experts LLM with ~314B total and 13.2B active parameters, trained from scratch on ~12.5T tokens with extra Korean, reasoning and agentic data. It uses Grouped Differential Latent Attention (GDLA), Expert-Specific PolyNorm, modified mHC and a built-in MTP head for self-speculative decoding. Native context is 262,144 tokens. It targets long-horizon agentic tool use and terminal tasks, with reasoning and tool-call parsers for vLLM. MIT License; base checkpoint Motif-Technologies/Motif-3-Base.

Providers

No active provider routes are tracked for this model yet.

Capabilities

ReasoningJSON / Tool use

Benchmark peer barsfor Coding

Benchmark scores(4)

Scores are benchmark-specific and are direction-aware: the same numeric gap can mean very different outcomes across suites. Use the leaderboard context and this model's provider route to decide whether the winning margin is meaningful for your workload.
BenchmarkScoreVersionEvaluationSource
76.2SWE-Bench VerifiedObserved 2026-10-09—
74.9Terminal-Bench 2.1Observed 2026-10-09—
83.4diamondObserved 2026-10-09—
37.0HLE (accuracy)Observed 2026-10-09—

Migration checks

No linked migration route is available for this model yet.

API versions

Motif-Technologies/Motif-3Motif 3